Abstract
This study tests whether AI-enabled personalized learning analytics (AI-PLA) improve EFL grammar acquisition in China and who benefits. A cross-sectional survey in four cities yielded N = 472 cases. AI-PLA quality was modeled as a higher-order construct; learner engagement and emotional intelligence (EI) served as mediator and moderator; grammar acquisition was assessed with the Oxford Placement Test Use-of-English score. SmartPLS 4 supported hierarchical modeling, a latent interaction, and diagnostics. AI-PLA quality directly predicted grammar (β = .19, p < .001) and strongly predicted engagement (β = .48, p < .001). Engagement predicted grammar (β = .41, p < .001) and mediated the AI-PLA → grammar link (β_indirect = .20, p < .001), indicating partial mediation. EI moderated the AI-PLA → engagement path (β_interaction = .12, p = .004). R2engagement = .56; R2grammar = .44. Theoretically, we link analytics quality—information quality, system quality, feedback timeliness, perceived personalization—to grammar outcomes through engagement and identify EI as a boundary condition. Practically, we recommend analytics that provide accurate error detection, feedback, usable interfaces, and visible personalization, alongside routines that cultivate engagement and emotion regulation.
Keywords
Introduction
The grammar of English as a Foreign Language (EFL) has also been an obstinate bottleneck to most learners since it requires not only the declarative knowledge of the forms but also a proceduralized control in the real-time application. With large-scale programs of EFL instruction in China, where accountability is measured by exams and where digitalization is accelerating, it becomes of interest whether AI-enabled personalized learning analytics (AI-PLA) can shift the needle on grammar gains by changing ubiquitous learner data into individually targeted feedback. Based on the information systems and second language (L2) engagement research, the study proposes and hypothesizes a model with AI-PLA quality as the independent variable, EFL grammar acquisition as the dependent variable, and learner engagement as the mediator with emotional intelligence (EI) as the moderator and cross-sectional and quantitative data and structural equation modeling estimated on SmartPLS (J. F. Hair, 2014).
AI-PLA refers to the bundle of system capabilities that collect, analyze, and visualize learner traces to drive adaptive recommendations and feedback loops. From an information systems perspective, the effectiveness of such tools hinges on the perceived quality of the system and the information it delivers (DeLone & McLean, 2003) and on the degree of meaningful personalization achieved (DeLone & McLean, 2003; Komiak & Benbasat, 2006). In EFL settings, analytics dashboards can surface actionable insights—error patterns with tense/aspect, article use, or preposition choice—and deliver just-in-time guidance that aligns with learners’ goals, thereby enhancing usability and utility (Rets et al., 2021). We conceptualize AI-PLA quality as a higher-order construct capturing information quality, system quality, analytics feedback quality/timeliness, and perceived personalization; the expectation is that higher AI-PLA quality will be associated with stronger grammar outcomes because learners receive clearer signals about what to practice and why (DeLone & McLean, 2003; Komiak & Benbasat, 2006; Rets et al., 2021).
To make this construct concrete in the present study, we use AI-enabled personalized learning analytics (AI-PLA) to mean learner-facing dashboards and reports that (a) aggregate clickstream and task-performance data from the grammar platform, (b) visualize progress at the level of specific grammar structures, and (c) generate individualized recommendations about which units to review next and which error types to prioritize. In other words, AI-PLA in this study is not a generic reference to “any” analytics, but to a specific bundle of information, system, feedback, and personalization features that students can inspect between and after classes.
However, analytics alone rarely translate into achievement without the motivational-cognitive engine of engagement. L2 engagement—comprising behavioral persistence, cognitive strategy use, and affective involvement—has emerged as a proximal driver of language progress because it channels attention, effort, and regulation toward difficult forms and structures (Harper & Quaye, 2009). We therefore posit learner engagement as a mediator linking AI-PLA quality to grammar acquisition: when analytics feedback is clear, timely, and personalized, learners are more likely to invest sustained, strategic effort in grammar practice, which in turn yields measurable gains. In operational terms, engagement captures the “work” that converts data-informed recommendations into deliberate practice cycles and feedback uptake (Harper & Quaye, 2009).
At the same time, not all learners respond equally to analytics and adaptive scaffolding. Emotional intelligence—typically modeled with the Wong and Law Emotional Intelligence Scale (WLEIS) covering self- and other-emotion appraisal, use of emotion, and regulation—can amplify or dampen the effects of AI-PLA on engagement because it shapes how learners interpret feedback, manage anxiety, and sustain motivation when confronting error-focused information (Kong et al., 2013; Shi & Wang, 2007; Wong & Law, 2002). Learners higher in EI may leverage analytics more effectively: they regulate frustration, maintain self-efficacy, and persist through challenging grammar tasks, thereby strengthening the AI-PLA → engagement pathway. Ultimately, grammar acquisition is assessed as performance, reflecting the integration of form-meaning-use knowledge (Purpura, 2004). By testing this mediated–moderated model among Chinese EFL students, the present study contributes theoretically by integrating IS quality and L2 engagement perspectives and offers practical guidance for deploying AI-PLA to convert data into durable grammar learning through emotionally intelligent, engaged participation (DeLone & McLean, 2003; Hiver et al., 2020).
Despite rapid growth in educational data mining and dashboard research, much of the literature on learning analytics in language education is descriptive or focuses on perceived usefulness and interface design, with limited attention to grammar-specific attainment and explanatory mechanisms (Rets et al., 2021). Studies that do examine outcomes typically aggregate language proficiency rather than isolating grammar, and they seldom connect analytics features to performance through a theoretically specified mediator (Hiver et al., 2020). This study addresses these gaps by modeling AI-enabled personalized learning analytics (AI-PLA) quality as the independent variable and EFL grammar acquisition as the dependent variable, while positing learner engagement as the mediating mechanism that converts analytics into deliberate practice and measurable gains. Grammar is assessed through performance-based indicators aligned with established assessment frameworks to avoid sole reliance on self-report (Purpura, 2004).
A second gap concerns the fragmented treatment of technology quality. Many investigations treat “learning analytics” as a monolith, overlooking core information systems dimensions—information quality, system quality, personalization, and feedback timeliness—that are theorized to drive effectiveness (DeLone & McLean, 2003; Komiak & Benbasat, 2006). This study explicitly models AI-PLA quality as a higher-order construct and estimates it with a two-stage PLS-SEM approach to capture its multidimensional nature and reduce construct misspecification (Becker et al., 2012; J. Hair & Alamer, 2022). By linking these IS quality facets to engagement and performance, the study integrates analytics design with learning processes rather than treating dashboards as a black box.
A third gap involves individual differences. Emotional processes are central when analytics highlight errors, yet the moderating role of emotional intelligence (EI) remains underexplored in EFL analytics, especially in China’s large, exam-oriented programs (Kong et al., 2013; Shi & Wang, 2007; Wong & Law, 2002). This study tests EI as a boundary condition on the AI-PLA → engagement pathway, clarifying for whom analytics are most effective. Methodologically, prior work often relies on covariance-based models with simpler structures; this project employs PLS-SEM to handle hierarchical constructs and interaction terms while emphasizing predictive relevance for grammar outcomes (J. Hair & Alamer, 2022). Collectively, the study advances theory by integrating IS success and L2 engagement perspectives, and it advances practice by specifying which analytics qualities matter, how they operate through engagement, and under which emotional profiles they yield the largest grammar gains.
At the same time, our goal is not to claim that grammar learning has been ignored in technology-enhanced language learning. A substantial body of computer-assisted language learning (CALL) and technology-supported language learning research has examined how digital tools, intelligent tutoring systems, mobile applications, and serious games foster grammar development and related linguistic competences (Bahari & Gholami, 2022; Sánchez Castro et al., 2024; Shadiev & Wang, 2022). What remains under-specified, however, is how learners interpret and respond to AI-driven analytics about their own grammar performance, and how these perceptions combine with engagement and emotional characteristics to shape objective grammar gains in regular EFL programs. The present study therefore narrows the gap to a more realistic but still important question: under what conditions do AI-PLA dashboards for grammar coincide with better test performance, and for which learners?
Guided by these gaps, we address five research questions (RQs) in the present study:
In order to bring out the logic of the study, a hypothesis test is provided to each research question. Particularly, H1, H2, H3, H4 (indirect effect/mediation), and H5 (interaction/moderation) address RQ1, RQ2, RQ3, RQ4, and RQ5 respectively. Therefore, the testing of the hypotheses gives direct empirical responses to the questions of the research.
Literature Survey
A growing strand of research argues that learning analytics can catalyze language learning when data are translated into timely, individualized feedback rather than static dashboards. In EFL contexts, analytics that highlight error patterns in tense/aspect, articles, and prepositions, coupled with targeted recommendations, are theorized to reduce ambiguity in study decisions and accelerate deliberate practice (Suphon, 2019). To ground this claim, information systems work emphasizes that technology effects are conditional on quality—specifically information quality (accuracy, relevance, completeness) and system quality (usability, reliability, responsiveness)—as well as the degree of meaningful personalization that learners perceive (DeLone & McLean, 2003; Komiak & Benbasat, 2006). Building on this foundation, the present study conceptualizes AI-enabled personalized learning analytics (AI-PLA) quality as a multidimensional, higher-order construct that includes information quality, system quality, analytics feedback quality/timeliness, and perceived personalization. Positioning AI-PLA quality as the independent variable reflects a shift from treating analytics as a monolith to theorizing the specific design attributes most likely to move grammar outcomes (En et al., 2025).
Converging evidence in applied linguistics shows that engagement is a proximal engine of achievement because it channels attention, effort, and strategy use toward difficult forms and tasks. L2 engagement is commonly modeled as behavioral (persistence and participation), cognitive (deep strategies and self-regulation), and affective (interest and enjoyment), and has been linked to higher quality practice and better performance across skills (Hiver et al., 2020; Teravainen-Goff, 2023). In AI-supported learning, dashboards and recommendation agents are best seen as “choice architectures” whose effects materialize only when learners act on them. Accordingly, this review supports a mediated pathway in which clear, timely, and personalized analytics increase behavioral, cognitive, and affective engagement, which in turn promotes grammar acquisition (Chiah Siew Fei, 2019). Conceptually, engagement is the mechanism that converts analytics into repeated, feedback-informed practice cycles.
At the same time, not all learners benefit equally from error-focused, data-rich environments. Emotional intelligence (EI)—typically operationalized with the Wong and Law Emotional Intelligence Scale (WLEIS) spanning self- and other-emotion appraisal, use of emotion, and regulation—shapes how students interpret feedback, cope with anxiety, and sustain motivation amid corrective signals (Kong et al., 2013; Shi & Wang, 2007; Wong & Law, 2002). Evidence from Chinese university samples supports the reliability and validity of WLEIS and links EI to adaptive coping and academic functioning, suggesting a plausible boundary condition on technology–engagement relations in EFL programs where high-stakes assessment can heighten negative affect (Kong et al., 2013; Shi & Wang, 2007). Framing EI as a moderator on the AI-PLA → engagement path therefore aligns with theory and measurement practice while acknowledging individual differences salient in China’s large, exam-oriented classrooms.
Finally, grammar acquisition is most defensibly treated as performance rather than perception. Assessment work highlights the value of aligning measures with the form–meaning–use triad and of aggregating items into reliable parcels or using externally validated subtests when available (Purpura, 2004). Bringing these strands together, the literature supports a mediated–moderated model: higher AI-PLA quality should predict stronger engagement; engagement should, in turn, predict better grammar performance; and EI should strengthen the first link by helping learners regulate affect and persist through analytics-driven correction. Methodologically, partial least squares structural equation modeling (PLS-SEM) is appropriate for estimating hierarchical constructs and interaction terms while emphasizing predictive relevance—an orientation congruent with the applied goal of improving grammar outcomes through analytics design (Becker et al., 2012; J. Hair & Alamer, 2022).
These trends have been supported and extended by more recent research on AI-enhanced personalized learning and learning analytics. The results of the research of personalized recommendation systems and dashboard AI-enabled systems are always positive, and they are associated with engagement, self-regulation, and achievement of students in the field of higher education and in language learning (Huang et al., 2023; Merino-Campos, 2025; Molla-Esparza et al., 2025; Vorobyeva et al., 2025). Parallel reviews in educational technology and CALL also highlight the fact that analytics-driven feedback can support grammar and more general language development if tightly linked to the course objectives and provided with training for learners in how to interpret visualizations (Chen, 2024; Kaur et al., 2023; Shadiev & Wang, 2022). Our paper places AI-PLA in grammar learning in this new body of evidence and pays particular attention to the issue of learner engagement and emotional traits.
Theoretical Foundation
The current research is based on the DeLone and McLean (2003) Information Systems (IS) Success Model, which assumes the influence of the perceived quality of information system and its outputs on the outcomes of the users. Quality, in the context of an AI-enabled personalized learning analytics (AI-PLA) is defensibly considered a multidimensional perception of what the learners get and experience when engaging with the analytics context. In line with this, we theorize AI-PLA quality as the perceptions of (a) information quality, (b) system quality, (c) feedback/service quality, and (d) personalization quality by the learners. It is assumed that these quality perceptions will go hand in hand with enhanced learner engagement since more transparent, more trustworthy, and more practical analytics cause less friction and make grammar-centered learning more objective (DeLone & McLean, 2003).
In this IS-success framing, the role of the learner as an intermediate process between perceived quality in analytics and learning outcomes is placed. Engagement elicits long term behavior investment, mental investment, and good participation in grammar-related study by the learners. In instances where AI-PLA is high quality, the learners have a better chance to stay attentive to recommendations, continue working through error-driven feedback, and convert analytics cues into purposeful practice, which ought to be associated with higher grammar acquisition achievements.
In order to state an empirically testable bound condition that is reflected in what we measure and analyze, we appeal to Control-Value Theory (CVT) of achievement emotions (Pekrun, 2006). CVT insists that situational contexts of achievement have the ability to trigger emotion-valenced appraisals (e.g., confidence, anxiety, frustration), thereby influencing engagement. Demonstrating an error or a gap in progress in the form of a dashboard in analytics-based grammar learning may invoke affective responses to reinforce or even discourage further action. Emotional intelligence (EI), or the ability to recognize, interpret, and manage feelings thus, should be a humble moderator: students with higher EI can be somewhat more efficient in remaining engaged in the situations when performance shortages are brought to the fore by analytics (Pekrun, 2006; Wong and Law, 2002).
Overall, the IS Success Model suggests the essential explanatory framework of why the perceived AI-PLA quality must be related to engagement and grammar acquisition and CVT allows EI to be a theoretically appropriate limit to this. Based on this framework, we test a directional (but cross-sectional and thus correlational) mediated-moderated model where the quality of perceived AI-PLA is directly and indirectly predictive of grammar acquisition through engagement, and EI moderates the direct AI-PLA quality engagement relationship.
Hypotheses
AI-enabled personalized learning analytics (AI-PLA) synthesize learner trace data into targeted recommendations and just-in-time feedback that can close persistent gaps in EFL grammar—such as tense/aspect, articles, and prepositions—by directing attention to high-value practice opportunities. From an information systems perspective, technology exerts effects on outcomes when its information quality (accuracy, relevance, completeness), system quality (usability, reliability, responsiveness), and personalization are salient to users (DeLone & McLean, 2003; Komiak & Benbasat, 2006). EFL analytics dashboards and recommendation agents that effectively visualize the patterns of errors and provide valuable corrective feedback in a timely fashion can help to shorten the feedback loop and reduce uncertainty about what to study next and scaffold effective rehearsal of the hard-to-learn grammar forms (Rets et al., 2021). This is particularly consequential in the big programs of China which are primarily exam-oriented where personalized feedback on teachers is minimal and students are left to wonder on their own between lessons.
There is a conceptual rationale, independent of motivational mechanisms, in having the direct positive link between AI-PLA and achievement. The high-quality analytics operationalize the enhancement of input and targeted output by (a) surfacing input contingencies of form/meaning/use that underlie grammatical competence and (b) directing intentional, criterion-based practice that brings morphosyntactic knowledge to procedural control (Purpura, 2004). With analytics that are correct, timely, and tailored, learners can devote study time to their most influential types of errors, they can be guided on corrective measures in real-time, and they can measure their progress in fine-grained terms: all of which are likely to produce measurable effects in grammar performance tests. In line with IS Success logic, such benefits in net form when the quality attributes of technology are strong enough to yield reliable outcome relevant signals (DeLone & McLean, 2003). Thus, the research hypothesis is that AI-PLA quality has a main-effect relationship to EFL grammar acquisition without (but complemented by) motivational processes.
The theorized form of high-quality AI-powered personalized learning analytics (AI-PLA) is believed to bring vitality to the engagement of learners by simplifying the process of study choices, accelerating feedback systems, and personalizing the significance of goals. Through the information systems perspective, the perceptions of the quality of information (accuracy, relevance, completeness), system quality (usability, reliability, responsiveness), and meaningful personalization are the proximal variables of user reaction to information systems including use and satisfaction (DeLone & McLean, 2003; Komiak & Benbasat, 2006). In language learning, they are the responses that affect the engagements in the form of sustained behavioral effort, more profound strategic processing, and positive affect when performing the tasks the three facets of engagement most directly related to the achievement (Hiver et al., 2020). Demonstrating the pattern of errors and featuring just-in-time and personalized recommendations, learning analytics dashboards minimize confusion regarding what is the next step to take, which is a primary precondition of persistence and strategic control (Rets et al., 2021). Self-Determination Theory also postulates that the contexts that promote competence and autonomy give rise to high-quality engagement; the analytics that offers diagnostic feedback, in addition to having mastery-oriented advice and a choice of practice paths directly address these needs (Deci & Ryan, 2000; Ryan et al., 2021). Within large, exam-based EFL programs in China, where individual feedback between teachers is limited, AI-PLA can replace granular cues (responses) that maintain focus and effort during inter-class time with granular, timely cues that proffer action during inter-class time as choice architectures. Aligning correct signals to individual pathways, AI-PLA enhances the chances of learners to invest in grammar work cognitively (use of strategies, self-monitoring), behaviorally (time-on-task, persistence), and affectively (interest, less frustration; Hiver et al., 2020; Rets et al., 2021). Therefore, the research will assume a direct motivation channel between the quality of technology and engagement.
Learner engagement is a proximal driver of language acquisition since it focuses attention, effort and strategic control into the specific forms and functions that support grammatical competence. Engagement (behavioral, time-on-task, persistence), cognitive (deep strategy use, self-monitoring), and affective (interest, enjoyment) aspects of engagement are generally described in L2 research; the consistent congruity of these aspects is always linked to improved-quality practice and performance (Harper & Quaye, 2009; Hiver et al., 2020; Teravainen-Goff, 2023). In grammar, in particular, the improvement is based on repeated cycles of intensive practice and feedback that enhance the mapping between form, meaning and use; active learners have a higher likelihood to (a) identify recurring patterns of errors, (b) implement corrective measures, and (c) persevere until procedural control is achieved (Purpura, 2004). Contexts that facilitate autonomy and competence, viewed through the motivational perspective, are likely to generate high-quality engagement, or deep processing and persistence, which, in turn, provides performance gains (Ryan et al., 2021). The emotion processes also play a role: learners put in more effort and regulate themselves better in case they have activating, controllable emotions (e.g., constructive challenge), which also increases the engagement-achievement relationship (Pekrun, 2017). Students in China with large, exam oriented EFL programs, where individualization is seldom provided through the teacher, are particularly determined to maintain deliberate feedback-based practice between classes; students who are behaviorally and cognitively engaged can more easily consolidate morphosyntactic knowledge and transfer it to tests. Integrating such strands, engagement is not an outcome product but the instant process in which study behaviors build up to proficiency.
Building on information systems and motivational theories, the relationship between AI-enabled personalized learning analytics (AI-PLA) and grammar outcomes is best understood as operating through learner engagement. The IS Success Model posits that information and system quality influence individual benefits primarily via proximal user responses (use, satisfaction), implying that technology quality exerts its effects through intermediary states rather than solely by direct transmission to outcomes (DeLone & McLean, 2003). In language learning, these proximal responses manifest as engagement—sustained behavioral effort, deep cognitive strategy use, and positive affect during tasks—rather than generic “use” (Hiver et al., 2020). When analytics are accurate, timely, and personalized, they clarify what to practice, reduce search costs, and supply diagnostic feedback, thereby satisfying competence and autonomy needs that energize high-quality engagement (Deci & Ryan, 2000; Ryan et al., 2021). Engagement, in turn, is the engine converting guidance into repeated, feedback-informed cycles of rehearsal that consolidate the form–meaning–use mappings underlying grammatical competence (Purpura, 2004). Empirically, recent work links stronger L2 engagement with deeper processing and better performance, suggesting that the benefits of technology-enabled interventions materialize through learners’ ongoing investment of attention and effort (Hiver et al., 2020; Teravainen-Goff, 2023). In China’s large, exam-oriented programs—where individualized teacher feedback is limited—such an engagement pathway is especially plausible: analytics can scale guidance, but only engaged learners translate signals into deliberate practice that yields measurable grammar gains. Accordingly, the model predicts a positive indirect effect from AI-PLA quality to EFL grammar acquisition via learner engagement.
When it highlights the instances of constant grammar errors, it may cause anxiety, frustration, or even shame, especially when it is used in the context of high-stakes coaching, such as in the Chinese EFL program. Control-Value Theory predicts that such emotions influence the engagement through the appraisals of control (can I fix this?) and value (does it matter?) by learners (Pekrun, 2017). Emotional intelligence (EI) also prepares students to evaluate feedback correctly, use activating emotions and down-regulate debilitating ones, maintain attention, strategic processing, and persistence against corrective signaling. Operationalized using Wong and Law Emotional Intelligence Scale (WLEIS), EI includes self- and other-emotion appraisal, use of emotion to facilitate performance, and emotion regulation, which have been established in Chinese university students and have been associated with adaptive coping and academic functioning (Kong et al., 2013; Shi & Wang, 2007; Wong & Law, 2002). Provided that the AI-enabled personalized learning analytics (AI-PLA) provide the correct, timely, and personalized feedback, high-EI learners are more prone to decoding the errors as active information, maintain motivation, and transform the recommendations into the intentional practice. On the other hand, lower EI learners have the threat appraisals that reduce engagement despite the high quality of analytics. Therefore, EI cannot simply increase engagement: it must enhance the beneficial impact of the quality of AI-PLA on engagement by converting the emotional connotation of feedback by switching the ego-threat emotion to the mastery challenge one. The model describes the heterogeneity in willingness to act on analytics as a result of technology quality by identifying EI as a boundary condition to the technology-engagement relationship. In this framework, emotional intelligence functions as a boundary condition on how learners respond to potentially ego-threatening information. Learners who can monitor and regulate their emotions are more likely to interpret red flags in dashboards as informative cues for adjustment rather than as global judgments of ability. They may thus remain engaged with AI-PLA even when confronted with frequent error signals, whereas learners with lower emotional intelligence may disengage or avoid analytics. Accordingly, we hypothesize a positive interaction between EI and AI-PLA quality in predicting engagement.
Methods
Research Site and Participants
We surveyed EFL learners enrolled in undergraduate English courses at four tertiary institutions in mainland China: two public universities, one vocational college, and one private language institute. After data screening (see Section “Sample Size Planning”), the final sample comprised 472 learners (58.9% female; see Table 1). Most participants were between 18 and 23 years old (45.3% aged 18–20, 44.5% aged 21–23), with a smaller group aged 24 or above. Year-of-study and major distributions reflected the mixed EFL provision at these institutions: 43.6% were English or applied linguistics majors and 56.4% were non-English majors taking compulsory EFL courses. Self-reported prior proficiency, benchmarked against CEFR levels, ranged from A2 to B2+, with A2 (25.0%), B1 (50.0%), and B2+ (25.0%) bands all represented. Weekly AI-PLA use also varied, with almost half of the sample reporting 1 to 3 hr/week and smaller groups at lower and higher usage bands. Together, this heterogeneity allowed us to examine AI-PLA, engagement, and grammar acquisition across a realistic cross-section of Chinese university EFL learners. Table 1 summarizes the socio-demographic profile and organizational segments. Briefly, the sample was geographically balanced across the four cities, skewed slightly female, and primarily at intermediate proficiency (B1). Most respondents were enrolled in public universities, with meaningful representation from private universities, vocational colleges, and private language institutes—mirroring the heterogeneous EFL delivery landscape examined in this study.
Sample Socio-Demographic Characteristics (N = 472).
Note. Percentages may not total 100 due to rounding.
An invitation was disseminated by course coordinators in every institution that characterized the study as an assessment of AI-assisted grammar study. In each of the participating classes, the first author or a trained research assistant came during regular EFL sessions, briefly outlined the study, and asked all students present to take part in the study on an anonymous basis. Students who were interested accessed the survey and grammar test using their own devices following a QR code and only a few filled out paper-based forms which were later scanned. It was voluntary and did not require any course credit or grade incentives.
Sample Size Planning
Because our structural model includes a reflective–reflective higher-order construct (AI-PLA quality) and a latent interaction term (AI-PLA quality × emotional intelligence), we followed recent sample-size recommendations for PLS-SEM models with hierarchical components and interactions (J. Hair & Alamer, 2022; Wang et al., 2025). With four first-order indicators of AI-PLA quality, one second-order engagement construct, and the interaction term, the maximum number of arrows pointing at a single endogenous construct was three. Monte Carlo simulations reported in the PLS-SEM literature suggest that, for small-to-medium effect sizes (f2 ≈ 0.05–0.15), a sample of around 400 to 450 cases yields power above 0.80 for such models; our N = 472 therefore provides adequate power for testing the hypothesized paths.
Instruments for Data Collection
All focal constructs were measured with established, validated scales and adapted for the present context using forward–back translation and cognitive interviewing. Items were anchored on five-point Likert scales (1 = strongly disagree, 5 = strongly agree) unless otherwise noted.
AI-PLA Quality
Perceived AI-PLA quality was operationalized through four first-order reflective dimensions—information quality, system quality, feedback quality/timeliness, and perceived personalization—adapted from information systems success and learning analytics research (DeLone & McLean, 2003; Rets et al., 2022; Susnjak et al., 2022). Each dimension was measured with 3 to 4 items capturing the clarity and usefulness of dashboard information (information quality), reliability and ease of use of the platform (system quality), timeliness and specificity of feedback on grammar tasks (feedback quality/timeliness), and the extent to which recommendations were tailored to learners’ weaknesses (perceived personalization). Table 2 reports reliability and validity statistics, and the second-order AI-PLA quality construct is modeled as a formative combination of these four dimensions.
Reliability, Validity, and Bias/Collinearity Diagnostics (N = 72).
Note.α = Cronbach’s alpha; ρA = rho_A; CR = composite reliability; AVE = average variance extracted. √AVE should exceed the construct’s highest correlation (“Max r with others”); HTMT<sub>max</sub> = highest heterotrait–monotrait ratio with any other construct (all < .85; 95% CIs <0.90); “Full coll. VIF” are Kock’s full collinearity VIFs used as an omnibus check for common method bias (all <3.3). The second-order AI-PLA quality construct was estimated using a disjoint two-stage approach with its four first-order reflective dimensions as formative indicators in stage two; reported are formative weights and their significance (*p < .05, **p < .01, ***p < .001) and formative VIF range. Fornell–Larcker criterion was assessed for reflective constructs (DV excluded because it is single-indicator). SRMR = .058; NFI = .91. All thresholds met or exceeded.
Learner Engagement
Learner engagement in grammar-focused study was measured with a multi-dimensional scale covering behavioral, cognitive, and emotional engagement, adapted from contemporary engagement research in EFL and higher education (Fredricks et al., 2019; Teravainen-Goff, 2023). Items tapped effortful participation in grammar tasks, strategic use of AI-PLA feedback, and affective investment (e.g., interest and persistence when facing analytic feedback about errors). As shown in Table 2, the three subscales functioned as reliable first-order factors that in turn loaded onto a single second-order engagement construct.
Emotional Intelligence
Trait emotional intelligence was assessed with the Wong and Law Emotional Intelligence Scale (WLEIS; Wong & Law, 2002), which has been validated for Chinese university students (Shi & Wang, 2007). Items asked learners to self-assess their ability to perceive, understand, and regulate emotions in themselves and others. We used the overall EI score as a moderator in the structural model, consistent with prior work on EI in academic contexts (Kong & Zhao, 2013).
Grammar Acquisition
Grammar acquisition was operationalized as performance on a timed, discrete-point grammar test aligned with the grammar units that participants were studying in their courses. The test consisted of multiple-choice and short sentence-completion items targeting intermediate morphosyntactic features that are central to Chinese university EFL curricula (e.g., tense–aspect contrasts, articles, relative clauses, complex verb phrases), developed with reference to principles for grammar assessment outlined by Purpura (2004). Items were written to avoid trivial recognition of isolated rules: most questions required integrating form and meaning in short contexts rather than filling in decontextualized blanks. A single total score (percentage correct) was used as the observed indicator of grammar acquisition (see Table 2).
Control Variables
To describe the sample and allow sensitivity checks, we also recorded learners’ self-reported CEFR band (A2, B1, B2+), gender, year of study, major, weekly AI-PLA use, organization type, and city (see Table 1). These variables were not the focus of the structural model but informed interpretation of the results and robustness checks. To mitigate common method bias, we used several procedural remedies: assuring participants of anonymity, separating the grammar test from the questionnaire section, mixing item blocks from different constructs, and including two instructed-response items. Analytically, we inspected full collinearity VIFs and a common-latent-factor model; all full collinearity VIFs were below 3.3 and the common-method factor did not account for the majority of variance in the indicators (Kock, 2015; Podsakoff et al., 2003). These checks reduce—but do not eliminate—the risk of common method bias, a limitation we return to in the “Discussion.”
Data collection process
To control the data collection among institutions, the protocols and instruments were identical over all the four sites within a 4-week period during the middle of the semester when students should have used the AI-PLA dashboard a few weeks earlier, but still worked on the target grammar units. Since the proficiency of the language may influence both the engagement and performance, the self-reported CEFR band A2, B1, B2+ (see Table 1) of learners were recorded, and the grammar test was developed to address intermediate constructions that are found in the core syllabi of each of the given levels. This minimized chances of advanced learners being bored by the test as well as lower-performing learners experiencing pervasive floor effects. In its interpretation, we consequently focus AI-PLA, engagement and grammar acquisition as acting in a context of moderate to a high degree of proficiency as opposed to extreme beginner or extreme advanced proficiency.
Data Analysis and Model Specification
We used partial least squares structural equation modeling (PLS-SEM) in SmartPLS 4 to estimate both the measurement and structural models. AI-PLA quality was modeled as a second-order formative construct composed of four first-order reflective dimensions: information quality, system quality, feedback quality/timeliness, and perceived personalization. Learner engagement was specified as a reflective higher-order factor indicated by behavioral, cognitive, and emotional engagement subscales. Emotional intelligence and grammar acquisition were modeled as first-order constructs, with grammar acquisition represented by a single observed test score. Following the two-stage approach for hierarchical component models, we first obtained latent-variable scores for the first-order constructs and then used these scores to estimate the higher-order AI-PLA quality construct and the latent interaction term (AI-PLA quality × emotional intelligence) via the product-indicator method.
All analyses were done using SmartPLS 4, which is highly appropriate with hierarchical component model (HCM) of the current study, latent interaction to moderate, non-normal indicators, and the objectives of the study are prediction-oriented. PLS-SEM allows disjointed two-stage estimation on our second-order AI-PLA construct, consistent PLS estimation on reflective blocks, and nonparametric bootstrapping of inference-strengths, which are consistent with our sample size and model complexity.
The structural model included a hypothesis test in each research question. The direct path of perceived AI-PLA quality to grammar acquisition tested H1 (RQ1). H2 (RQ2) was tested using the direct route of perceived AI-PLA quality to engagement of the learner. The H3 (RQ3) hypothesis was tested using the route between learning engagement and acquisition of grammar. H4 (RQ4) was tested using the bootstrapped indirect effect of the quality of AI-PLA and grammar acquisition through engagement. H5 (RQ5) was tested through interaction term (AI-PLA quality × emotional intelligence) to predict interaction, and then simple-slope probing at ±1 SD of EI.
Ethics
The guidelines of Helsinki Declarations were followed. The protocol received institutional ethics approval from Academic Ethics Committee of the host university. Participation was voluntary with no course credit or monetary incentives. Before the first item, participants read an informed consent statement describing risks, benefits, confidentiality, data retention (encrypted storage; de-identified exports), and the right to withdraw at any time without penalty. Only age (to confirm ≥18), city, institution type, and study-related variables were recorded. Under-18 students were not eligible.
Results
All analyses were conducted in SmartPLS 4, which is well suited to this study’s hierarchical component model (HCM), the latent interaction for moderation, non-normal indicators, and the study’s prediction-oriented goals. PLS-SEM enables disjoint two-stage estimation for our second-order AI-PLA construct, uses consistent PLS for reflective blocks, and provides nonparametric bootstrapping for inference—advantages that align with our sample size and model complexity.
Fitness of the Data
Data screening on the 505 returns produced N = 472 valid cases after removing submissions that failed attention checks, exhibited extreme straight-lining, had implausibly short completion times, or exceeded missing-data thresholds. Reflective measurement quality was strong: all standardized loadings exceeded .70, item reliabilities exceeded .50, and internal consistency indices were high (Cronbach’s α, ρA, and composite reliability ranged .87–.95). Convergent validity was established with AVE ≥ .64 for all reflective constructs. Discriminant validity held by both Fornell–Larcker (√AVE for each construct exceeded its largest inter-construct correlation) and HTMT (all HTMT ratios <.85; 95% bootstrap CIs stayed below 0.90). The second-order formative AI-PLA construct showed significant indicator weights for its four first-order dimensions with acceptable collinearity (VIFs < 2.6).
Evidence against common method bias was multi-pronged. Procedurally, we used temporal separation for the objective OPT grammar score, anonymity, mixed scale formats, and attention checks. Statistically, Harman’s single-factor accounted for 33.7% of variance, the full collinearity VIFs for all latent variables were <3.3 (range 1.8–2.8), and a common latent factor captured ≤24% additional variance; together these suggest CMB is unlikely to bias estimates. Multicollinearity was not a concern at either the outer- or inner-model level (outer VIFs < 3.0; inner VIFs for endogenous blocks ≤2.7). Model fit indices were acceptable for PLS-SEM: SRMR = .058 and NFI = .91, supporting adequate model-implied correlations.
Table 2 summarizes reliability, validity, and collinearity diagnostics. In brief, all reflective constructs met or exceeded conventional thresholds (α/CR ≥ .70; AVE ≥ .50; HTMT < .85), the second-order formative AI-PLA exhibited significant weights with low collinearity, and omnibus CMB screens were negative—collectively indicating a healthy measurement model suitable for hypothesis testing.
Hypotheses Testing
Structural paths were estimated in SmartPLS 4 using disjoint two-stage modeling for the higher-order AI-PLA construct and 5,000 bias-corrected bootstrap resamples (two-tailed; see Table 3). Standardized coefficients (β), t values, p values, and 95% BCa confidence intervals are reported. Inner-VIFs were ≤2.7, indicating no multicollinearity issues. The model explained substantial variance in Learner Engagement (R2 = .56) and Grammar Acquisition (R2 = .44), supporting the predictive focus of PLS-SEM.
Bootstrapped Path Estimates and Hypothesis Decisions (N = 472; 5,000 Resamples).
Note. BCa = bias-corrected and accelerated. Mediation evaluated via bootstrapped indirect effect (H4). Interaction probed via simple slopes at ±1 SD of EI (H5).
H1 (AI-PLA → Grammar Acquisition)
Perceived AI-PLA quality showed a small but significant positive association with grammar acquisition (β = .19, t = 4.02, p < .001, f2 = 0.05), indicating that learners who evaluated the dashboard as higher quality tended, on average, to achieve higher grammar test scores. Given the cross-sectional design, this association should be interpreted as correlational rather than causal.
H2 (AI-PLA → Learner Engagement)
AI-PLA quality was strongly associated with learner engagement (β = .48, t = 11.30, p < .001, f2 = .38). Learners who perceived the analytics as informative, reliable, and personalized reported substantially higher levels of behavioral, cognitive, and emotional engagement with grammar study.
H3 (Learner Engagement → Grammar Acquisition)
Engagement, in turn, was moderately associated with grammar acquisition (β = .41, t = 8.20, p < .001, f2 = 0.21), suggesting that more engaged learners tended to obtain better scores on the grammar test.
H4 (Mediation: AI-PLA → Engagement → Grammar Acquisition)
The indirect effect of AI-PLA quality on grammar acquisition via engagement was significant (β_indirect = .20, t = 6.40, p < .001), while the direct AI-PLA → grammar path remained positive but reduced in magnitude. This pattern is consistent with partial mediation: AI-PLA quality is related to grammar outcomes both directly and through its association with engagement.
H5 (Moderation: EI on AI-PLA → Engagement)
The interaction between AI-PLA quality and emotional intelligence in predicting engagement was statistically significant but small in magnitude (β_interaction = .12, t = 2.88, p = .004, f2 = 0.02). Simple-slope probes indicated that the AI-PLA–engagement association was slightly stronger for learners one standard deviation above the mean in EI than for those one standard deviation below. Thus, EI appears to be a modest boundary condition on how learners respond to AI-PLA rather than a dramatic moderator.
Answer to Research Questions
RQ1 was whether perceived AI-PLA quality is tied to EFL learners’ grammar acquisition. In accordance with H1, there was a small positive statistically significant relationship between perceived AI-PLA quality and grammar acquisition (0.19, 4.02, p < .001).
RQ2 was whether learner engagement is related to perceived AI-PLA quality. In keeping with H2, the quality of AI-PLA was highly related to the engagement of the learners (= 0.48, t = 11.30, and p = .001).
RQ3 was a question of whether the engagement of learners is a predictor of grammar acquisition in this setting of AI-PLA. In line with H3, grammar acquisition was positively related to engagement (0.41, t = 8.20, p < .001).
RQ4 was whether the relationship between the quality of AI-PLA and grammar acquisition is mediated by engagement. In line with H4, the bootstrapped indirect effect was also significant (0.20, t = 6.40, p < .001), which shows that engagement explains a significant part of the AI-PLA quality grammar relationship.
The fifth research question wanted to know whether emotional intelligence influences AI-PLA quality to engagement relationship. In line with H5, the interaction term was significant but the magnitude was low (β_interaction = 0.12, t = 2.88, p = .004) which corresponds to a low degree of strengthening the association between AI-PLA quality and engagement at increased levels of EI.
Discussion
This paper has explored the relationship between perceptions of EFL learners on AI-powered personalized learning analytics in the context of engagement and grammar learning in Chinese universities. Expanding on the findings above, the discussion descriptions interpret the pattern of associations, present practical implications of AI-PLA design and instructional application and comments on limitations that limit causal assertions.
Combined, the findings suggest a consistent yet empirically minor trend in line with an information-systems success view: learners that perceive AI-PLA quality co-are also associated with better engagement and, less extensively, better grammar acquisition results. The mediation result indicates that the most effective way in which AI-PLA quality is connected to grammar performance is via engagement. Moreover, EI acts as a small boundary condition which, to some degree, reinforces the quality-engagement relationship among learners capable of controlling achievement-related emotions that the analytics feedback may induce. Since the study is cross-sectional in nature and some of the constructs are perception-based, these results should be treated as directional associations as opposed to causes.
Theoretical Implications
The results are mostly an extension of the previous studies as they apply the existing models to the grammar-learning setting with the support of AI and PLA instead of creating a completely new theory. In line with the IS Success Model, perceived quality of AI-PLA among learners was similar and more so associated with greater engagement and, to a lesser degree, with better grammar results (DeLone & McLean, 2003). The outcomes of the mediation indicate the practical importance of considering engagement as an intermediary that learners can use to transfer analytics perceptions to learning outcomes. Secondly, the insignificance of the interaction effect indicates that emotional abilities can act as a weak boundary condition in analytics-intensive setting, which is congruent with the focus of CVT in processes of engagement related with emotion (Pekrun, 2006). On the whole, the research provides a logical empirical example of these connections in EFL grammar learning, and the longitudinal or experimental studies are required to be able to test the causality.
Practical Implications
Practically, the results offer cautious guidance for instructors, program leaders, and platform designers. First, because perceived information, feedback, and personalization qualities are strongly associated with engagement, it seems worthwhile to invest in dashboards that clearly visualize grammar progress at the feature level and that provide specific, timely recommendations for practice. Second, the small EI moderation effect suggests that simple emotion-aware routines—such as normalizing errors, encouraging reflective rather than punitive interpretations of analytics, or offering optional emotion-regulation tips—may help some learners stay engaged, though our data do not speak to the effectiveness of any particular training program. Importantly, without behavioral trace data on how learners actually clicked through dashboards or changed their study plans, we cannot recommend specific “choice architecture” features. Our recommendations should therefore be read as design hypotheses to be tested in future experimental or mixed-methods work rather than as direct prescriptions.
Limitations and Potential Research
This study has several limitations that qualify the conclusions. First, the design is cross-sectional and the questionnaire and grammar test were administered in a single sitting, separated only by a short break. Although PLS-SEM encourages directional modeling grounded in theory, the observed paths are correlational; we cannot rule out reverse or reciprocal influences (e.g., learners who are already strong in grammar may view AI-PLA more positively). Second, while we implemented multiple procedural and statistical remedies, common method bias cannot be definitively excluded because AI-PLA quality, engagement, and emotional intelligence were all self-reported in the same session. Third, our AI-PLA quality construct captures learners’ perceptions of information, system, feedback, and personalization quality rather than any objective system logs or algorithmic performance metrics. Conclusions about “high-quality analytics” should thus be understood as statements about students’ experiences with analytics, not about the underlying technology itself.
Future research should therefore combine perception measures with behavioral and system-level data, such as trace logs of how learners navigate dashboards, which recommendations they follow, and how their grammar performance evolves over time. Longitudinal or experimental designs that manipulate dashboard features, feedback framing, or emotion-regulation supports would allow stronger causal inferences about how AI-PLA, engagement, and emotions jointly shape grammar development. Qualitative work could also explore how learners interpret analytics visualizations and how teachers integrate AI-PLA into grammar instruction, extending the present focus on learners in mainstream Chinese tertiary EFL programs.
Conclusion
This study examined how learners’ perceptions of AI-enabled personalized learning analytics relate to engagement and grammar acquisition in Chinese university EFL courses. Perceived AI-PLA quality was strongly associated with engagement and modestly associated with grammar test performance, with engagement partially mediating this link and emotional intelligence providing a small moderating effect. While these patterns fit well with IS Success, SDT, and CVT perspectives, they should be interpreted cautiously given the cross-sectional design, single-session data collection, and reliance on self-reports for key predictors. Even so, the results underline a practical message: analytics appear most useful for grammar learning when learners experience them as clear, reliable, and personalized and when pedagogical routines help students stay engaged with feedback rather than overwhelmed by it.
Footnotes
Ethical Considerations
The guidelines of Helsinki Declarations were followed. The protocol received institutional ethics approval from Academic Ethics Committee of Harbin Normal University China, under approval code HNU-WLC-2024 1019 APL. Written informed consents were obtained from each responded prior to data collection.
Consent to Participate
Informed consent was obtained from all subjects involved in the study.
Author Contributions
All of the authors contributed to conceptualization, formal analysis, investigation, methodology, and writing and editing of the original draft
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
Data will be available on request from the corresponding author.